The Achaemenid dynasty

The thirteen kings ran from Cyrus II to Darius III. The reigns are fixed with unusual precision for so remote a period, because the empire's Babylonian subjects dated their contracts and their astronomical records by the regnal years of the reigning king, and enough of those dated tablets survive to anchor the chronology almost year by year against the absolute dates of recorded eclipses; the later Greek chronographic tradition, above all the king-list preserved through Ptolemy's Canon, agrees with them.^[1] The boundary years where one reign passes to the next are given as single years for clarity; several fell across the turn of the Babylonian year, and where a later reckoning shifts one by a year it is dual-dated here (405/4, 359/8), as the scholarship also cites such years (424/3).

Three of the thirteen barely reigned at all. The man who held the throne in 522 as Bardiya, Cyrus's younger son, and whom Darius branded the impostor magus Gaumāta (see Bardiya and Gaumāta), lasted about six months. In the succession crisis of 424 two half-brothers, Xerxes II and Sogdianus, held power for a few weeks and a few months respectively and were never recognised in Babylonia at all; and Arses, set on the throne by the eunuch Bagoas in 338, was murdered by the same hand two years later. The list also records a change in royal naming that begins in the same period: from Darius II onward the kings ruled under a throne-name different from the one they were born with, so that a Darius or an Artaxerxes might have begun life as an Ochus, an Arsaces or an Artašata. Ochus's adoption of the name Darius on his accession in 424 is the first clearly attested Achaemenid use of a throne-name, though the practice, familiar from the Neo-Assyrian court, where the designated crown prince took a new name, may be older and simply invisible in the earlier evidence.^[2][3]

The dynasty is usually called Achaemenid, after Achaemenes; but that name papers over a real division, and to see it one has only to set the two founding genealogies side by side. They do not describe the same house.

The older of the two is Cyrus's. On the Cyrus Cylinder, written at Babylon in 539, and on his monuments at Pasargadae, Cyrus gives his descent without a trace of Achaemenes. He is, in the Cylinder's words, "son of Cambyses, great king, king of Anshan, grandson of Cyrus, great king, king of Anshan, descendant of Teispes, great king, king of Anshan".^[1] The pedigree descends from Teispes through Cyrus I and Cambyses I to Cyrus II, every man of it styled king of Anshan, the old highland Elamite kingdom whose territory the Persians of Fārs had settled and whose royal style the early Teispids took over. The title was a claim in itself. By calling themselves kings of Anshan the founders presented their rule as continuous with Elam rather than as a Persian novelty, and the Persian administration Cyrus and Cambyses left behind, the one Darius inherited, was itself run in Elamite; Henkelman has made the point sharply by titling the whole problem a case of "mistaken identity", the supposedly straightforward opposition of a Persian Cyrus and a Persian Darius dissolving on inspection into something far more entangled.^[4] What is not in doubt is that "king of Anshan" and "Achaemenid" are titles from two different registers, and that only the first is demonstrably Cyrus's own. He is a Teispid, of the line of Teispes; he is never, on any document of his own, an Achaemenid.

One small object corroborates the Teispid pedigree from outside the royal monuments. A seal inscribed for "Cyrus, the Anshanite, son of Teispes", presumably the founder's grandfather Cyrus I, went on being used to stamp tablets in the Persepolis administration under Darius I, a prestige item, perhaps an heirloom, that shows the Anshanite royal line was a physical memory at the Persian court two generations after Darius had renamed the family.^[3]

The second genealogy is Darius's, and it enters the record only after his violent accession in 522. In the great inscription at Behistun Darius sets out a pedigree that reaches back past his father and grandfather to a new and senior name:

Darius the king proclaims: My father is Vishtaspa; Vishtaspa's father is Arshama; Arshama's father is Ariaramna (Gr. Ariaramnes); Ariaramna's father is Cishpish (Gr. Teispes); Cishpish's father is Hakhaimanish (Gr. Achaemenes).

Darius I, DB §2^[1]

From this chain Darius draws the family's name and its antiquity: "For this reason we are called Achaemenids. From long ago we are noble; from long ago we are royal," and, more boldly, "Eight of our family were kings before; I am the ninth; nine kings are we in succession."^[1] The genealogy does two things at once. It supplies the family with an eponym, Achaemenes, from whom the name Achaemenid is then derived; and it makes Darius the ninth king of a line reaching far into the past. Yet he never names those eight predecessors, and his own father Hystaspes and grandfather Arsames, both demonstrably alive in 522, had never been kings at all: Xerxes says as much of them in his own inscription, that "both Hystaspes and Arsames were living" when Darius came to the throne.^[5] The only way to make Darius's list and Cyrus's agree is to suppose that Teispes had two sons and split his kingdom in two, a senior branch running to Cyrus and Cambyses and a junior one running through Ariaramnes and Arsames to Hystaspes and Darius. On that reconstruction Darius and Cyrus are second cousins twice over, joined at Teispes. Everything turns on whether the split is real.

The strongest sceptical reading, argued at length by Pierre Briant and developed by Matt Waters, is that the split is not real, and that Darius manufactured the whole connection. Its evidence is partly what the sources say and partly what they conspicuously do not.

Briant presses the internal difficulties first. If Teispes really divided his kingdom, the eight kings before Darius cannot be found: Darius "is careful not to name them", and his father and grandfather, alive and untitled in 522, plainly were not among them. Nothing outside Darius's own claim attests a junior Teispid line; the separate inscriptions in which Ariaramnes and Arsames style themselves Great King are, on grounds of language and formula, later fabrications. "Despite Herodotus (VII.11)," Briant concludes, "the theory of two Persian kingdoms will not stand."^[6] The Greek tradition, read closely, corroborates the doubt rather than the claim: Plato and Aelian both state that Darius was not the son of a king, Justin implies the same, and Herodotus, who calls the pre-522 Darius a man "of no power or consequence", gives no hint that he had any better right to the throne than his six fellow conspirators.^[6] What Darius wanted, on this reading, was a new foundation for his own rule; so he "chose an eponymous ancestor, Achaemenes", by whom the kingship could be confined to one narrow line of descent, and around whom a founder-legend could be built. The redefinition did double duty. Achaemenid had been the name of a wide clan; by narrowing it to his own patriline Darius shut even his fellow conspirators out of the succession, so that, in Briant's summary, "power could only be transmitted from father to son in one restricted family, which was placed under the blessing and protection of a founder-hero, Achaemenes, invented out of whole cloth".^[6]

A second, philological support for the two-branch theory has since fallen away. Darius's phrase in DB §4, once rendered "nine kings in two lines", turns on the Old Persian word duvitāparanam, and the "two lines" reading was long taken to license the split. Rüdiger Schmitt's edition of the inscription and Robert Rollinger's philological studies both reject it: Schmitt renders the word "now as ever", and Rollinger argues that translating it "in two lines" is not possible. On the current understanding the inscription does not describe two branches at all; it claims one unbroken succession, and the branching is a modern harmonisation imposed to save the appearance.^[7]

The most concrete evidence for manufacture comes from Pasargadae. Two short inscriptions there, CMa and CMc (the latter's Old Persian version assumed), call the founder an Achaemenid: CMa reads simply, "I am Cyrus the King, an Achaemenid." On their face they prove Cyrus used the name; but Darius claimed in the same Behistun text to have invented Old Persian cuneiform, which no genuine inscription of Cyrus could then be written in, and Cyrus's authentic Babylonian titles never once say Achaemenid, only king of Anshan. Waters draws the conclusion that the Pasargadae texts are all suspect, added in Cyrus's name after the fact: "Darius thus retroactively transformed the illustrious Cyrus into one of his own ancestors and strengthened his own legitimacy."^[3] Amélie Kuhrt reaches the same point from the shape of the Behistun genealogy itself, whose "resulting awkwardness" betrays, in her judgement, a pedigree manipulated to manufacture "an indisputable family-right to the throne": Darius was "almost certainly a member of the extended clan" of the Persian kings, "but his relationship to Cyrus' family was probably not particularly close."^[7] The genealogy was built in the aftermath of a violent seizure of power and served to legitimate it (see the accession of Darius).

The sceptical case is now the majority view, but it has been complicated, and in one respect blunted, by the documentary turn. Wouter Henkelman, working from the Persepolis Fortification archive, has argued that the sharp opposition of a Teispid and an Achaemenid house, of Cyrus against Darius, is itself partly a modern artefact, inherited from Herodotus's antithesis of Cyrus the father and Darius the shopkeeper. "In its strict form," he writes, "the traditional contrast between Cyrus and Darius is not corroborated by primary evidence."^[4] Darius was probably playing on the double sense of Old Persian taumā, which means both a wide clan and a narrow ruling family, but that he belonged to the Persian ruling elite is not in doubt.

What the archive adds is that Darius did not treat Cyrus's family as a rival to be erased. "It appears that Darius was certainly not pursuing a policy of damnatio memoriae vis-à-vis the Teispids," Henkelman notes: the Fortification tablets record funerary offerings maintained on the king's orders for Cambyses, performed by court-titled chamberlains and organised, the dossier shows, so as "to ensure the regularity and continuity of the sacrifices"; and the surviving members of Cyrus's house held great estates and moved freely at the centre of the empire. Atossa and Artystone, daughters of Cyrus, became Darius's wives; Artystone was, after the king and the great lady Irdabama, the wealthiest individual named in the whole archive, and her son Arsames acted as the recognised head of the Teispid branch. "All this suggests that the remaining Teispids acted as a group and were recognised as such," and were not shut out.^[4] On this reading the genealogy is less a lie than a merger: Darius did not deny Cyrus's line but absorbed it, by marriage and by cult, into a single house under a name of his own choosing.

On this much, in fact, the sceptics agree. Briant, for all the bluntness of "whole cloth", holds that Darius never sought to efface the founder's memory; at Pasargadae, rather, "he intended to promote skillful propaganda … that would allow him to establish a fictitious link with Cyrus, just as he did with his matrimonial policy".^[6] The disagreement between the sceptical and the archive-led readings is therefore narrower than it first appears: both have Darius deliberately attaching himself to Cyrus's house rather than replacing it. What divides them is whether any kinship at all lay behind the attachment.

A minority keeps open the possibility that the connection was real: Bruno Jacobs has argued that Cyrus may have been an Achaemenid after all, though a distant relative of Darius and one of many with an equal right.^[4] The genealogy Herodotus gives Xerxes at 7.11, with its two Teispes-figures, does not quite match the Behistun list; Kuhrt reads it not as an independent witness but as an echo of Darius's genealogical amalgam.^[7] The evidence does not force a single answer. What it will not support is the old picture of a seamless dynasty descending unproblematically from Achaemenes to Darius III.

The field has moved, without reaching consensus. The reconstruction of two houses, Teispid and Achaemenid, joined by Darius, is now standard enough that recent reference works and the specialist literature routinely distinguish the Teispid kings, Cyrus and Cambyses, from the Achaemenids proper, Darius and his successors, and treat the eponymous Achaemenes as Darius's contribution. The disagreement that remains is over what kind of act the joining was: an outright fabrication, on Briant's and Waters's reading, or the political consolidation of a genuine if distant kinship, on Henkelman's. What would settle it does not survive. A pre-Darius inscription of Cyrus, or of any of the eight unnamed kings, naming Achaemenes, would prove the connection old; its continued absence, at Pasargadae above all, is the strongest single argument that the connection is Darius's work. Until such a text appears, the dynasty's own foundation-story has to be read as a claim, and dated to the moment the claim was made.

The tree below draws the royal house as the tradition, once Darius had shaped it, held it to be: a single descent from Achaemenes, forking at Teispes into the line of Cyrus and the line of Darius, and rejoined in the next generation by the marriage of Darius I to Cyrus's daughter Atossa, whose son Xerxes therefore carried both bloods. From Xerxes the succession runs, for the most part directly, down the centre of the tree to the fourth century; the last king, Darius III, was a collateral, brought to the throne only when the direct line had been murdered away.

The queens whose place in the descent is secure and consequential are Cassandane, wife of Cyrus and mother of his sons; Atossa, the hinge of the whole tree; Amestris, wife of Xerxes; Damaspia, whose death on the same day as her husband Artaxerxes I opened the crisis of 424; Parysatis, the mother of Artaxerxes II; and Stateira. Behind these named few stood many more wives and, especially in the later reigns, a great number of sons by secondary unions, from whom the Greek writers spun their tales of harem intrigue.

For most of its history the crown passed from father to son, but the Persians had no simple rule of primogeniture, and the sources show that the reigning king designated his successor. Where a designation was clear the transfer could be smooth; where it was contested or absent, the result was a fight. The one principle the sources articulate is that a son born after his father became king outranked an elder son born before, in the Roman phrase born in the purple, though Kuhrt notes it is otherwise unattested as a Persian custom and it surfaces in only one succession dispute.^[5]

The clearest case is Xerxes' own. He was not Darius's eldest son; Artobazanes, by an earlier wife, was older. In Herodotus's account the two disputed the succession, Artobazanes urging that "it was customary among all mankind for the eldest to have precedence", Xerxes that he alone "had been born when Darius was already king and held power in Persia, while Artobazanes was born when Darius was just a private individual".^[5] In Herodotus that born-in-the-purple argument is supplied by the exiled Spartan Demaratus, and the decision goes, in the end, to Atossa's influence; Kuhrt judges the story "imaginatively elaborated", but its core, that Xerxes was not the eldest son and that his appointment was unusual, is confirmed by Xerxes himself, who states in his own inscription that "Darius had other sons also" and yet "made me the greatest", by the will of Ahura Mazdā.^[5] Designation, not birth order, made the king, and a king could reach past an elder son to a younger one born to a more prestigious mother.

The designation did not always hold. Three times the transfer broke down and had to be settled by force. In 522 the throne was seized in the name of Cambyses's brother Bardiya while Cambyses still lived; Cambyses then died with no recognised heir, and Darius and the six conspirators killed the sitting king. In 424 Artaxerxes I and his queen Damaspia died on the same day, and their only legitimate son, Xerxes II, was murdered within some forty-five days by his half-brother Sogdianus, who was himself overthrown within months by a third half-brother, Ochus, ruling as Darius II; none of the ephemeral pair was ever recognised in the Babylonian dating.^[2] In 338 the death of Artaxerxes III, which later tradition blames on the courtier Bagoas, began a second forced sequence that ran through the boy-king Arses to the collateral Darius III. Between these crises the principle worked: designation by the reigning king, weighted toward the son of the most royal wife, carried the crown down a single line for generations at a stretch. The Greek image of the Persian court as a place of permanent murderous intrigue takes the exceptions for the rule.

The royal women mattered to the dynasty in two ways the Greek sources tend to run together and distort. The first was strictly dynastic. Marriage was how the house secured and concentrated its bloodline: Darius married two daughters of Cyrus and a daughter of Bardiya, tying his new kingship to the founder's line, and the succession of his son by Atossa turned in part on her descent from Cyrus. The marriages did in blood what the Behistun genealogy did in words: where descent from the founder could be asserted, it was also, for safety, married in.^[9] Brosius argues that behind such choices lay a rule of endogamy, that "a woman had to be Persian to become a king's wife, in order to provide a Persian heir to the throne", which is why foreign princesses, however grand, entered the court as concubines rather than as queens.^[9]

The second was the trope of the harem, and here the sources need handling with care. The figure of three hundred or three hundred and sixty royal concubines comes from hostile fourth-century writers, Heracleides, Deinon and Plutarch, whose aim was to give "an exaggerated and distorted view of the Persian kings and to present them as weak and effeminate"; Herodotus, earlier and less prurient, gives the palace women no such influence.^[10] The very existence of a segregated harem quarter at Persepolis rests on a nineteenth- and twentieth-century identification, of Herzfeld's and Schmidt's, of one palace building as the women's wing, which Margaret Cool Root cautioned against, its position on a low level of the terrace and its rows of small compartments seeming to her to disprove the identification.^[11][10] What the documents do attest, in the Fortification tablets rather than the Greek gossip, is that royal women held estates, commanded large workforces and travelled with their own establishments: their real and documented power was economic and administrative, not the bedroom intrigue of the Greek portrait (see Atossa).^[4]

Around the direct line of kings ran a wider royal kindred, and the empire made systematic use of it. Younger sons and brothers of the king, and the descendants of earlier kings, were regularly given the greatest satrapies and the highest commands, so that a Cyrus the Younger could govern the whole of Anatolia for his brother Artaxerxes II, and princes of the blood turn up as satraps, generals and admirals throughout the two centuries. This was a source of strength and of danger at once: it bound the provinces to the house, but it also meant that a disappointed prince had the resources to rebel, as Cyrus the Younger did in 401.

What is striking about even the ugliest of these contests is how narrow the field remained. In the scramble after Artaxerxes I's death in 424, Waters observes, "all claimants had one thing in common: direct descent from the Achaemenid line": the sons of a Babylonian secondary wife could fight one another for the throne, but no one outside the blood royal entered the competition at all.^[3] The winner, Ochus, then chose to reign as Darius, a throne-name that tied him to the dynasty's second founder; "Ochus could not have done much more to solidify his dynastic credentials."^[3] A century after Darius I had narrowed the royal taumā to his own descendants, the narrowing still held: the dynasty's quarrels were family quarrels, fought entirely inside the fence Darius had built.

The collateral lines also supplied the throne when the direct line failed. Darius III, the last king, was not of the senior stem at all: born about 380, he was a great-grandson of Darius II through a younger son, Ostanes, and reached the crown only because Bagoas's murders had cleared the direct descent away.^[12]

The two centuries fall into three acts. The first is the founding, the work of the Teispid kings: within a single generation Cyrus overthrew the Median king, the Lydian Croesus, and Babylon, and his son Cambyses II conquered Egypt, so that by 522 the largest empire the world had yet seen was in Persian hands and barely twenty years old. The second act opens with the crisis of that year: a man ruling as Cambyses's brother Bardiya took the throne, Cambyses died without a clear heir, and Darius, at the head of six fellow conspirators, killed the sitting king and seized the crown. It was Darius who gave the empire its lasting shape, its satrapies and tribute, its roads and coinage, and Darius and his son Xerxes who carried it to its limits and, in the invasion of Greece, to its first great check.

The third act is the long fourth century, less famous and worse recorded, in which the empire proved far more durable than the Greek picture of decline allows. The century is punctuated by disputed successions, the murderous one of 424 and the war between Artaxerxes II and his brother Cyrus the Younger that ended at Cunaxa in 401, and by revolts of over-mighty satraps; yet the centre held, and Artaxerxes III late in the period recovered Egypt and restored the empire's reach. The notion of a slow "Achaemenid decadence" is a Greek and, above all, a Xenophontic construct, and Briant has argued that the evidence will not bear it: on the eve of Alexander's invasion "Darius III's Empire was not moribund, as it was smugly described by the Greco-Hellenistic authors", and "there is nothing to suggest that, from Darius I to Darius III, the internal capabilities of the Empire were crumbling". The theory of decline, he concludes, "must definitively be relegated to a display case in the museum of historiographic wonders".^[13]

The dynasty was brought down not by rot but by conquest, and by a last convulsion at its own centre. The murders of 338 to 335 had already thinned the direct line to nothing: as Diodorus tells it, when Bagoas killed Arses and his children "the royal house was thus extinguished, and there was no one in the direct line of descent to claim the throne" — a claim the surviving evidence overstates, as Artaxerxes IV sets out — so that the kingmaker had to reach into a collateral branch and set up Darius III, who promptly had Bagoas drink his own poison.^[14] Within six years of Alexander's landing in 334 the last Darius had lost three great battles and, in 330, his life, murdered in flight by one of his own satraps; with him the dynasty of Cyrus and Darius came to an end.

Its prestige outlasted it. Alexander took Darius III's daughter Stateira to wife and presented himself, in Persian eyes, as the legitimate heir to the Great Kings; and for centuries afterward the local dynasts of Anatolia and the Iranian world, the kings of Pontus and Cappadocia and, most elaborately, Antiochus I of Commagene, traced their pedigrees back to Darius and to the Achaemenid house, staking a claim to a legitimacy that the name still carried long after the last king who held it had died. The Achaemenid descent had become, by then, the most valuable ancestry in the Near East.